Climate-driven chemistry and aerosol feedbacks in CMIP6 Earth system models

Abstract

Abstract. Feedbacks play a fundamental role in determining the magnitude of the response of the climate system to external forcing, such as from anthropogenic emissions. The latest generation of Earth system models includes aerosol and chemistry components that interact with each other and with the biosphere. These interactions introduce a complex web of feedbacks that is important to understand and quantify. This paper addresses multiple pathways for aerosol and chemical feedbacks in Earth system models. These focus on changes in natural emissions (dust, sea salt, dimethyl sulfide, biogenic volatile organic compounds (BVOCs) and lightning) and changes in reaction rates for methane and ozone chemistry. The feedback terms are then given by the sensitivity of a pathway to climate change multiplied by the radiative effect of the change. We find that the overall climate feedback through chemistry and aerosols is negative in the sixth Coupled Model Intercomparison Project (CMIP6) Earth system models due to increased negative forcing from aerosols in a climate with warmer surface temperatures following a quadrupling of CO2 concentrations. This is principally due to increased emissions of sea salt and BVOCs which are sensitive to climate change and cause strong negative radiative forcings. Increased chemical loss of ozone and methane also contributes to a negative feedback. However, overall methane lifetime is expected to increase in a warmer climate due to increased BVOCs. Increased emissions of methane from wetlands would also offset some of the negative feedbacks. The CMIP6 experimental design did not allow the methane lifetime or methane emission changes to affect climate, so we found a robust negative contribution from interactive aerosols and chemistry to climate sensitivity in CMIP6 Earth system models.

Identifying the sources of uncertainty in climate model simulations of solar radiation modification with the G6sulfur and G6solar Geoengineering Model Intercomparison Project (GeoMIP) simulations

“…From the perspective of assessing ecosystem impacts, this decoupling of precipitation, temperatures and CO 2 should be investigated in depth to understand if and where it would be beneficial or not. It also further stresses the notion that reducing precipitation is not an automatic result of geoengineering but that the outcome is related to which specific cooling targets geoengineering is deployed to achieve (Tilmes et al, 2013;Irvine et al, 2019;Lee et al, 2020). All models agree, to various degrees, that global precipitation changes under G6sulfur are larger than the same changes under G6solar.…”

“…Further information on the models' components can be found in the references provided for each model and a summary is given in Table S1 in the Supplement. More detailed information for CMIP6 models can also be found in Séférian et al (2020) for marine biogeochemistry, Arora et al (2020) for carbon-climate feedbacks and Thornhill et al (2021) for atmospheric chemistry.…”

“…Similarly, it has been widely assessed that trying to restore surface temperature to a previous state by means of modifying the top of the atmosphere radiative balance tends to overcompensate the changes in precipitation, therefore reducing global mean precipitation. Globally, the changes are driven by the perturbation of the surface heat fluxes (Tilmes et al, 2013;Kravitz et al, 2013b;Niemeier et al, 2013) and changes in sea-land temperature contrast. Regionally, however, the modification of the baseline distribution of precipitation can be due to changes in the inter-tropical convergence zone (ITCZ; Russotto and Ackerman, 2018b;Cheng et al, 2019) produced by changes in the inter-hemispheric temperature gradient, general circulation changes produced by stratospheric heating (Simpson et al, 2019) and regional and seasonal changes in heat fluxes and temperature gradients (Jones et al, 2018;Visioni et al, 2020b).…”

“…Two previous experiments in particular have been widely analyzed and discussed: G1, where the solar constant is reduced in order to offset the temperature increase produced by a 4× increase in CO 2 compared to pre-industrial concentrations (Kravitz et al, 2013b;Tilmes et al, 2013;Glienke et al, 2015;Russotto and Ackerman, 2018b;Kravitz et al, 2021), and G4, where a constant amount of SO 2 is injected into the equatorial stratosphere under emissions from the Representative Concentration Pathway 4.5 (RCP4.5) (Pitari et al, 2014;Kashimura et al, 2017;Visioni et al, 2017b;Plazzotta et al, 2019. However, previously performed GeoMIP experiments were not intended to be "realistic" deployments of geoengineering, either because they were performed under idealized conditions (such as 4×CO 2 concentrations) or because they considered a fixed, constant amount of injected SO 2 with an abrupt beginning and ending.…”

“…In the stratosphere, these include the conversion of injected SO 2 into stratospheric aerosol and the subsequent large-scale distribution of the aerosols by stratospheric circulation (not dissimilar to multi-model analyses of simulations of explosive volcanic eruptions; Marshall et al, 2018;Clyne et al, 2021), the chemical response of key stratospheric components (ozone, methane) to the aerosol layer (Pitari et al, 2014;Visioni et al, 2017b), the magnitude of the produced local heating (Niemeier et al, 2020) and the dynamical response. At the surface, uncertainties include the magnitude of the resulting global cooling per Tg-SO 2 injected or per unit of optical depth produced, the regional patterns of change in temperature (Kravitz et al, 2013a), precipitation (Kravitz et al, 2013b;Tilmes et al, 2013) and extreme events (Aswathy et al, 2015;Ji et al, 2018) as well as other variables that might affect ecosystems and populations (Zarnetske et al, 2021), such as tropospheric ozone (Xia et al, 2017) or cloud changes (Russotto and Ackerman, 2018a).…”